Batteries: To fuse or not to fuse (or circuit breakers)

[mike90045]

So I was asked today, and thought I've give my view, and my setup that was county inspected 2 years ago.

I have a single string of batteries, which wire via Conduit, to the Midnight Solar ePanel
http://www.solar-electric.com/midnite.html
. The ePanel is a prefab wire box, with all the buss bars, rails, knockouts and air vents, designed to fit the DC & AC load points of large inverter systems.

Dead center of the ePanel is a 250A DC (really DC rated) breaker. There is about a 5' length of 4/0 cable between the inverter and battery bank.  The Charge controller is also wired to the same breaker, there is no fuse in the system, breakers only.

If there were 2 battery banks, each would have to have it's own fuse or breaker, so if a bank went really bad, both banks would not dump into the fault. here's a page full of high amp DC rated parts: http://www.solar-electric.com/cipr1.html

I hope the pics come out...

[Lloyd]

Mike,

It boils down to cost. A true DC breaker of that ampere will cost more than a Class T fuse, to replace.

If the breaker trips do to over current fault it should be replaced. The costs of a Class T of the same ampere rating will be about 1/3 the costs of a true dc breaker the ampere.

The OCPD should be sized 10 x's the bat bank amp hr for lead acid, and 20 x's the amp hr rating of a gel or agm bank. If it's a Li-bat  or equal the OCPD needs to be 100 x's amp hr rating of the bank.

OCPD = Over Current Protection Device.

Lloyd

[mike90045]

Midnight's ePanel for the XW, comes with breakers installed. Wasn't a choice.  Midnight generally has pretty robust parts

http://www.midnitesolar.com/productPhoto.php?product_ID=72&productCatName=Xantrex%20XW,%20DR/TR/Samlex%20E-Panels&productCat_ID=6
Features: inverter battery breaker, 2 separate 60 amp AC inputs for generator and utility, 60 amp AC bypass switch, knockouts for up to seven din rail mount DC, DCGFP63 or AC breakers and 12 panel mount or DC-GFP80 breakers. Massive tin plated copper bus bars directly connect to the XW's battery terminals, bus bars for AC inputs, AC output, neutral, ground, PV + in, PV- in, Bat +, Bat-, 500 amp shunt, metal dead front behind the reversible door, Charge controllers mount to either side or both sides at once for dual controllers. AC bypass can be configured as input & output on/off as well as AC bypass. The XW MPPT controller requires no mounting bracket. FM60, FM80 and Classic require Right or Left E-Panel charge control mounting brackets. Right hand bracket furnished standard.

250A DC breaker:
http://www.midnitesolar.com/productPhoto.php?product_ID=202&productCatName=Breakers&productCat_ID=16 

[WStayton]

Hi Guys!

  Clarification please!

  Lloyd said: "The OCPD should be sized 10 x's the bat bank amp hr for lead acid . . ."

  Does this mean that if I have a string of 225 amp hour batteries in series for a 48 VDC system, I need to have a 2,225 amp breaker/fuse?  If so, where do I buy such an animal???

  Also, 2,225 amps seem like it is only going to protect against dropping a crescent wrench across the terminals - not the 150%-200% overloads that are introduced because of some other problem!

  Or, do I not understand what is being said her . . . AGAIN!  <grin>

  Enlighten me, please!

Regardz,

Wayne Stayton

[Lloyd]

Wayne,

An OCPD, has two ratings.

1. Nominal rating, ie 400 amp, which is sized to protect the wire, not the load from to much current causing heating of said wire.

2. Fault Current rating, ie dead short, which is sized to the potential of the fault current.

Batteries can generate 10 to 100 times the nominal current into a dead short.

Mike,

No matter how robust a breaker is, one single direct fault in a dc OCPD can and will damage the breaker, as the breaker is 3 times the cost of the fuse. A fuse is cheaper, and should be installed as close to the battery as possible, the breaker in the panel is used as a service disconnect.

Never close a dc breaker under load, nor open a breaker into a full load, as this will also damage the breaker.

Lloyd

[rcavictim]

My 48 volt battery bank is split into two identical banks.  Each bank is 9 strings of four AGM, Valve regulated, 100 AH 12 volt batteries in series.  Each 48 volt series string of four batteries is protected by a 35 amp DC rated breaker to the master copper bussbar.  This way I can isolate individual strings of batteries for diagnosis, or I can load test them one string at a time.  The copper bus that represents the 9 strings is tied to a 400 A DC rated 175 volt Heinmann breaker.  There are two of these master 400 A breakers, each representing half of the total 18 string battery bank. The two big breakers feed a master inverter buss. I can close both and use all 72 batteries or use half of the bank while charging the other half.

I will only be able to charge or discharge my batteries at max 1/3rd C as a result of the 35 amp breaker limitation.  I will have outstanding fault protection however.  I do the best I can with the parts I am able to acquire within a modest budget.

Cannot post pics at this time due to a computer failure.  I have not started to use these batteies yet as my wind turbine is not yet finished.  I have been keeping them in shape applying occasional, ~monthly float supply for a day or so.

I still need to design and fabricate a 10-20 kW battery charger and charge controller.  It would be sweet to be able to just go and buy one.

[Tom Reed]

Wow it's busy in there!

[WStayton]

Lloyd:

  So, for my TWO strings of 48 volts (collectively), 225 amp hour rated batteries, what current rating fuses should I have on each string?

  Supposedly, the Xantrex inverter/charger will only output 60 amps continuous at 120 volts with a short term (think seconds) transient to twice that.  I'm thinking that I should have something like 180 or 200 amps on each string of batteries.  Yes, this is overkill, since, if I'm am using both strings (my intention) they will each be producing only half of the full load, but I want to have the capability of disconnecting each string occasionally, overnight, to be able to test them in a rested condition, and I don't want to be in the dark, doing this with a flashlight! <grin>

  So, will I be okay/all-right/in-the-ballpark with a 200 amp fuse on each string, or should I be going for half of that and just hoping that I don't have any large loads when I'm down to one string for testing?

  I suppose I could simply have two sets of fuses, one of 100's and one of 200's and then usual the 100's for normal, day-to-day, both string operation and insert the 200's when I am going to leave one string to rest for testing, but I fear that I will get lazy and not insert the 200's and/or lose/misplace them.

  On a slightly different note:  Is there something magic about DC fuse holders?  I note that they are at least twice the price of an AC fuse holder from HomeDepot and many are three or four times as much.  Can I just buy the appropriately current rated AC "box", slip the DC fuses in it and be good-to-go?

  Thanx for any input, guys - all is appreciated!

Regardz,

Wayne Stayton

[Tom Reed]

Wayne, The outback power center I have came with a 175 amp breaker for each 3600 watt 48vdc inverter. I believe you are going to want to size the breaker based on the ampacity of the cables you're installing.

[Lloyd]

Lloyd:

  So, for my TWO strings of 48 volts (collectively), 225 amp hour rated batteries, what current rating fuses should I have on each string?

  Supposedly, the Xantrex inverter/charger will only output 60 amps continuous at 120 volts with a short term (think seconds) transient to twice that.  I'm thinking that I should have something like 180 or 200 amps on each string of batteries.  Yes, this is overkill, since, if I'm am using both strings (my intention) they will each be producing only half of the full load, but I want to have the capability of disconnecting each string occasionally, overnight, to be able to test them in a rested condition, and I don't want to be in the dark, doing this with a flashlight! <grin>

  So, will I be okay/all-right/in-the-ballpark with a 200 amp fuse on each string, or should I be going for half of that and just hoping that I don't have any large loads when I'm down to one string for testing?

  I suppose I could simply have two sets of fuses, one of 100's and one of 200's and then usual the 100's for normal, day-to-day, both string operation and insert the 200's when I am going to leave one string to rest for testing, but I fear that I will get lazy and not insert the 200's and/or lose/misplace them.

  On a slightly different note:  Is there something magic about DC fuse holders?  I note that they are at least twice the price of an AC fuse holder from HomeDepot and many are three or four times as much.  Can I just buy the appropriately current rated AC "box", slip the DC fuses in it and be good-to-go?

  Thanx for any input, guys - all is appreciated!

Regardz,

Wayne Stayton

1. DC Fused at the battery to protect the cable to the load.

2. DC Breaker at the inverter likely won't protect the inverter from an dead short prior to, it will trip out if an over amp situation occurs on the downstream side  of the inverter. But those individual loads also should have their own breaker. So really  the breaker before/at the inverter trips out for an internal inverter fault, or provides a service disconnect when work to the inverter needs to happen. It should also have a service disconnect on the AC side of the inverter for the same reason. An inverter is a live circuit until you disconnect both AC, and DC.

3. DC fuse at the battery is rated at 125% the nominal amp load, and fault protection based on the amp hr size of the bank. Cable is sized to 150% rated ampere load.

4. AC breakers/fuse and cable are rated at by the same equation, bc AC is alternating current the protection devises are different.

5. Battery banks in parallel perform best with the tails of the bank that lead to the bus coming from opposite ends of the bank. While you can lead each separate bank to a common bus to make a parallel connection, one bank will always work harder then the other, and cell imbalance will be a bigger problem.

Lloyd

[WStayton]

Lloyd:

  You said:  "Battery banks in parallel perform best with the tails of the bank that lead to the bus coming from opposite ends of the bank. While you can lead each separate bank to a common bus to make a parallel connection, one bank will always work harder then the other, and cell imbalance will be a bigger problem."

   Huh???

  If I have two strings of eight each, six volt batteries, each with its own 100 amp fuse, tied in to a common "bus" which is wired to the Xantrex inverter/charger, how do I arrange them so that the "tails of the bank that lead to the bus coming from opposite ends of the bank."

  It seems to me, that if I arrange them so that they have the same lengths of cable in each leg, they should have the same resistance, and hence the same charge/discharge characteristics and nothing further is required, no?

  Enlighten me, I'm confused again, as usual!   <grin>

Regardz,

Wayne Stayton

[WStayton]

Let me repeat this question so that it doesn't get lost in the mess of my drivel!  <grin>

  Is there something magic/different about a fused DC disconnect that makes it inherently different from an AC disconnect?

  I ask, because DC disconnects are several times as expensive as AC disconnects, and I would think that if you had the right fuses in the disconnect (i.e. DC) and it was rated for that current, it wouldn't matter what you put through it.  Right?

  Or, is the differance that a DC disconnect can/may have only one pole while an AC disconnect normally has two poles, but still is way less money?

  Oh, and one other question:  If I do use two pole, AC fused disconnects, can I just use two of them, one for each string of batteries, and then wire the two poles in parallell and use two fuses that are each half of the total rating for the "leg"  Or, I could use just one half of the disconnect and then use the other "side" to store the larger fuse for use when only one "string" is in use.

  I'm about to plop down my money and I hate to get the wrong thing and then, with a red face, go back and ask for a refund/exchange!  <smile>

  So is there an inherent difference between a DC and an AC fused disconnect, other than the fuses?

  Inquireing minds want to know, and mine too!!! <grin>

Regardz,

Wayne Stayton

[cognos]

If I remember correctly, there is indeed a design difference between a DC disconnect and an AC disconnect.

About 33 years ago, for a short time, I sold motor control for Allen Bradley. If what's left of my mind serves me correctly in this instance, it is more difficult to open or close a DC circuit under load due to the DC's propensity to sustain an arc... hence, the contacts in a DC disconnect must open much faster and farther, have arc shields, be more robustly-built in order to take the abuse, etc... and this explains the extra cost of the device.

I know you shouldn't open or close a DC disconnect under load - but since it's a disconnnect, it has to be designed to be able to do so safely and reliably.

Hey, I could be wrong. But this this thread tickled my memory.

[Lloyd]

If I remember correctly, there is indeed a design difference between a DC disconnect and an AC disconnect.

About 33 years ago, for a short time, I sold motor control for Allen Bradley. If what's left of my mind serves me correctly in this instance, it is more difficult to open or close a DC circuit under load due to the DC's propensity to sustain an arc... hence, the contacts in a DC disconnect must open much faster and farther, have arc shields, be more robustly-built in order to take the abuse, etc... and this explains the extra cost of the device.

I know you shouldn't open or close a DC disconnect under load - but since it's a disconnnect, it has to be designed to be able to do so safely and reliably.

Hey, I could be wrong. But this this thread tickled my memory.

Cognos you are correct, also to note that if using a breaker especially in high amp dc loads the contacts are also of different material. High am DC generates a lot of heat, when opening/closing into a full load. Remember we use High Amp DC for welding.

The same thing happens to contacts when open or closed into a load, a dead fault will make reuse of the breaker unsafe for it's intended use, so must be replaced after such an event. That's why it's cheaper to use DC Rated fusing for high amp loads.

Wayne,

No you can't use AC fuses for HA DC loads, the fuse holder and the fuse must be UL listed for DC, and they must be rated to the ampere, and the potential fault.

I have provided at least 2 links below, that cite the codes. If you're going to build your system without the aid of a consultant, then you really need to spend sometime reading those links. It would be foolish for you to install your system without the full understanding of the codes.

While we can help you understand certain issues, we are not on location, to make sure what you understand is what we said.

Lloyd

[WStayton]

Lloyd said:  "While we can help you understand certain issues, we are not on location, to make sure what you understand is what we said."

  So since I don't have an electrical engineering degree I should just forget it?

  The fact that I don't have an electrical engineering degree is why I am on here asking - now your telling me that is bad practice???

  I'm confused - I THOUGHT this was a Do-It-Yourself forum - Am I wrong AGAIN??

  If I HAD an electrical engineering degree, I wouldn't be on here asking so many questions! <grin>

  Oh, and having watched the NYSEG engineer who DOES have an electrical engineering degre operate, and I'm pretty convinced that, left to his own devices, he would make bigger messes than I will!

  My opinion(s) which are worth EXACTLY what you paid for them!

Regardz,

Wayne Stayton